深度學習

Lightweight Flutter enterprise development skill focused on feature-based clean architecture. Use when user asks to: (1) Build enterprise Flutter apps with clean architecture, (2) Implement feature-based modular structure, (3) Set up scalable Flutter project organization, (4) Create maintainable enterprise codebase. Triggers: Flutter enterprise app, clean architecture Flutter, feature-based Flutter, enterprise Flutter structure, modular Flutter architecture

This skill should be used when creating, editing, or debugging Mermaid diagrams. It provides comprehensive syntax references for all 20+ Mermaid diagram types including flowcharts, sequence diagrams, class diagrams, ER diagrams, Gantt charts, and more. Use this skill when users ask for diagrams, visualizations, flowcharts, or any Mermaid-related assistance.

Add tensor shape validation and documentation to PyTorch code. This skill should be used when working with PyTorch models to ensure tensor shapes are correct and well-documented with inline comments.

Specialized agent for creating, maintaining, and organizing comprehensive documentation for the BarqNet project. Records all changes, creates technical specifications, API documentation, user guides, deployment guides, and maintains documentation consistency across the entire codebase. Use when documenting features, creating guides, or organizing project knowledge.

Breaks down specs into complexity-stratified tasks. Use when organizing tasks, assigning work to agents, or when user mentions task layering, task stratification, or parallel task execution.

Convert Python loops to vectorized PyTorch tensor operations for performance. This skill should be used when optimizing computational bottlenecks in PyTorch code during Phase 4 performance optimization.

Social return on investment (SROI) modeling, blended finance structuring, and impact-weighted accounting. Use for evaluating social enterprises, impact investments, ESG integration, sustainability metrics, and theory of change financial modeling. Generates comprehensive impact assessments that blend financial returns with social and environmental outcomes, suitable for impact investors, foundations, and mission-driven organizations.

Comprehensive GitHub Pull Request code review skill. This skill should be used when users provide a GitHub PR URL and request a code review. Automatically fetches PR metadata, diff, comments, commits, and related issues using gh CLI. Creates organized review workspace, analyzes code against industry-standard criteria (functionality, security, testing, maintainability), and optionally adds inline comments to the PR. Trigger phrases include review this PR, code review, review pull request, check this PR, or when a GitHub PR URL is provided.

Transforms workflow to use Manus-style persistent markdown files for planning, progress tracking, and knowledge storage. Use when starting complex tasks, multi-step projects, research tasks, or when the user mentions planning, organizing work, tracking progress, or wants structured output.

WHEN: Machine Learning/Deep Learning code review, PyTorch/TensorFlow patterns, Model training optimization, MLOps checksWHAT: Model architecture review + Training patterns + Data pipeline checks + GPU optimization + Experiment trackingWHEN NOT: Data analysis only → python-data-reviewer, General Python → python-reviewer

Master AI engineering with deep learning, TensorFlow, PyTorch, model deployment, and AI applications including LLMs and agents.

Use this skill whenever the user wants to design, set up, or refactor Cloudflare R2 object storage usage in a Hono + TypeScript app running on Cloudflare Workers/Pages, including bucket bindings, upload/download flows, signed URLs, and folder-like organization.

Templates and scripts for generating layered BUILD-GUIDE.json/md from Airtable plugin index - shows available commands organized by infrastructure layers

Validate PubMed search field tags against official NCBI documentation before implementing them. Use when adding new field tags to the query builder, debugging search queries, or verifying existing field tag implementations. Prevents invalid tags like [organism] and ensures compliance with NCBI PubMed standards.

Use when assessing team resilience, planning for developer departures, calculating bus/truck factor, identifying knowledge silos, or evaluating organizational risk - maps code ownership from git history and identifies single points of failure using research-backed thresholds (>80% ownership = silo)

Generate comprehensive Pest test suites for Laravel projects including tests for controllers, services, Livewire components, commands, jobs, and models with proper describe blocks for organization, mocking setup, happy path and failure scenarios, RefreshDatabase trait, and coverage for all public methods. Use this skill when creating test files for new features, adding test coverage for controllers or services, writing tests for API endpoints, testing Livewire components, ensuring proper test structure with beforeEach and describe blocks, implementing mocking strategies, or writing dataset-driven tests. This skill applies when working on files in tests/Feature/ or tests/Unit/ directories, creating Pest test files, or ensuring comprehensive test coverage.

Manage a personal knowledge capture system. Use when the user wants to capture ideas, track todos, organize projects, review their brain dump, or mentions "brain dump", "capture this", "add to my list", "what's on my plate", "what should I focus on", or "daily review".

Follow development best practices for project organization, version control, documentation, environment configuration, and dependency management to maintain consistency across the team. Use this skill when organizing project files and directory structures, writing or updating README files and documentation, working with git commits and branches, configuring environment variables, managing dependencies in package.json or requirements.txt, or establishing code review processes. Use this skill when working with configuration files, .env files, documentation files, or when setting up CI/CD workflows. Use this skill when implementing feature flags, maintaining changelogs, or establishing team development workflows and contribution guidelines.

Model fine-tuning with PyTorch and HuggingFace Trainer. Covers datasetpreparation, tokenization, training loops, TrainingArguments, SFTTrainerfor instruction tuning, evaluation, and checkpoint management.

Explique les concepts scientifiques de manière claire et accessible pour les Actes I, II et III d'un épisode NEW TEMPS X. Utilise ce Skill quand tu dois expliquer des concepts scientifiques, créer une introduction intrigante, décrypter un mécanisme scientifique, ou poser des questions ouvertes sur un sujet scientifique.